data_designer.config.models

Bases: str, enum.Enum

Supported modality types for multimodal model data.

Initialization:

Initialize self. See help(type(self)) for accurate signature.

1
IMAGE = image

Bases: str, enum.Enum

Data type formats for multimodal data.

Initialization:

Initialize self. See help(type(self)) for accurate signature.

1
URL = url

1
BASE64 = base64

Bases: str, enum.Enum

Types of distributions for sampling inference parameters.

Initialization:

Initialize self. See help(type(self)) for accurate signature.

1
UNIFORM = uniform

1
MANUAL = manual

Bases: abc.ABC, pydantic.BaseModel

1
modality: data_designer.config.models.Modality

1
column_name: str

1
data_type: data_designer.config.models.ModalityDataType | None

1
get_contexts(
2
    record: dict,
3
    *,
4
    base_path: str | None = None
5
) -> list[dict[str, typing.Any]]

Bases: data_designer.config.models.ModalityContext

Configuration for providing image context to multimodal models.

Parameters:

Attributes:

Initialization:

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

1
modality: data_designer.config.models.Modality

1
image_format: data_designer.config.utils.image_helpers.ImageFormat | None

1
get_contexts(
2
    record: dict,
3
    *,
4
    base_path: str | None = None
5
) -> list[dict[str, typing.Any]]

1
_auto_resolve_context_value(
2
    context_value: str,
3
    base_path: str | None
4
) -> dict[str, str]

1
_format_base64_context(base64_data: str) -> dict[str, str]

1
_validate_image_format() -> typing_extensions.Self

Bases: abc.ABC, data_designer.config.base.ConfigBase, typing.Generic[data_designer.config.models.DistributionParamsT]

1
distribution_type: data_designer.config.models.DistributionType

1
params: data_designer.config.models.DistributionParamsT

1
sample() -> float

Bases: data_designer.config.base.ConfigBase

Parameters for manual distribution sampling.

Parameters:

Attributes:

Initialization:

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

1
values: list[float] = Field(...)

1
weights: list[float] | None

1
_normalize_weights() -> typing_extensions.Self

1
_validate_equal_lengths() -> typing_extensions.Self

Bases: data_designer.config.models.Distribution[data_designer.config.models.ManualDistributionParams]

Manual (discrete) distribution for sampling inference parameters.

Samples from a discrete set of values with optional weights. Useful for testing specific values or creating custom probability distributions for temperature or top_p.

Attributes:

1
distribution_type: data_designer.config.models.DistributionType | None = manual

1
params: data_designer.config.models.ManualDistributionParams

1
sample() -> float

Bases: data_designer.config.base.ConfigBase

Parameters for uniform distribution sampling.

Parameters:

Attributes:

Initialization:

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

1
low: float

1
high: float

1
_validate_low_lt_high() -> typing_extensions.Self

Bases: data_designer.config.models.Distribution[data_designer.config.models.UniformDistributionParams]

Uniform distribution for sampling inference parameters.

Samples values uniformly between low and high bounds. Useful for exploring a continuous range of values for temperature or top_p.

Attributes:

1
distribution_type: data_designer.config.models.DistributionType | None = uniform

1
params: data_designer.config.models.UniformDistributionParams

1
sample() -> float

Bases: str, enum.Enum

1
CHAT_COMPLETION = chat-completion

1
EMBEDDING = embedding

1
IMAGE = image

Bases: data_designer.config.base.ConfigBase, abc.ABC

Base configuration for inference parameters.

Parameters:

Attributes:

Initialization:

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

1
generation_type: data_designer.config.models.GenerationType

1
max_parallel_requests: int = Field(...)

1
timeout: int | None = Field(...)

1
extra_body: dict[str, typing.Any] | None

1
generate_kwargs: dict[str, typing.Any]

1
format_for_display() -> str

1
get_formatted_params() -> list[str]

1
_format_value(
2
    key: str,
3
    value: typing.Any
4
) -> str

Bases: data_designer.config.models.BaseInferenceParams

Configuration for LLM inference parameters.

Parameters:

Attributes:

Initialization:

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

1
generation_type: typing.Literal[data_designer.config.models.GenerationType]

1
temperature: float | data_designer.config.models.DistributionT | None

1
top_p: float | data_designer.config.models.DistributionT | None

1
max_tokens: int | None = Field(...)

1
generate_kwargs: dict[str, typing.Any]

1
_validate_temperature() -> typing_extensions.Self

1
_validate_top_p() -> typing_extensions.Self

1
_run_validation(
2
    value: float | data_designer.config.models.DistributionT | None,
3
    param_name: str,
4
    min_value: float,
5
    max_value: float
6
) -> typing_extensions.Self

1
_is_value_in_range(
2
    value: float,
3
    min_value: float,
4
    max_value: float
5
) -> bool

1
_format_value(
2
    key: str,
3
    value: typing.Any
4
) -> str

Bases: data_designer.config.models.BaseInferenceParams

Configuration for embedding generation parameters.

Parameters:

Attributes:

Initialization:

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

1
generation_type: typing.Literal[data_designer.config.models.GenerationType]

1
encoding_format: typing.Literal[float, base64] = float

1
dimensions: int | None

1
generate_kwargs: dict[str, float | int]

Bases: data_designer.config.models.BaseInferenceParams

Configuration for image generation models.

Works for both diffusion and autoregressive image generation models. Pass all model-specific image options via extra_body.

Parameters:

Attributes:

Example:

1
# OpenAI-style (DALL·E): quality and size in extra_body or as top-level kwargs
2
dd.ImageInferenceParams(
3
    extra_body={"size": "1024x1024", "quality": "hd"}
4
)
5
6
# Gemini-style: generationConfig.imageConfig
7
dd.ImageInferenceParams(
8
    extra_body={
9
        "generationConfig": {
10
            "imageConfig": {
11
                "aspectRatio": "1:1",
12
                "imageSize": "1024"
13
            }
14
        }
15
    }
16
)

Initialization:

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

1
generation_type: typing.Literal[data_designer.config.models.GenerationType]

Bases: data_designer.config.base.ConfigBase

Configuration for a model used for generation.

Parameters:

Attributes:

Initialization:

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

1
alias: str

1
model: str

1
inference_parameters: data_designer.config.models.InferenceParamsT = Field(...)

1
provider: str | None

1
skip_health_check: bool = False

1
generation_type: data_designer.config.models.GenerationType

1
_convert_inference_parameters(value: typing.Any) -> typing.Any

1
_warn_on_implicit_provider() -> typing_extensions.Self

Bases: data_designer.config.base.ConfigBase

Configuration for a custom model provider.

Parameters:

Attributes:

Initialization:

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

1
name: str

1
endpoint: str

1
provider_type: str = openai

1
api_key: str | None

1
extra_body: dict[str, typing.Any] | None

1
extra_headers: dict[str, str] | None

1
normalize_provider_type(v: str) -> str